nraptisss
/

tmf921-intent-training

intent-based-networking

network-slicing

Model card Files Files and versions

tmf921-intent-training / scripts /train_rft.py

Commit History

Fix RFT: batch generation, reduce to 200 prompts x 8 samples (~24h feasible on RTX 6000 Ada)

05ea6fa
verified

nraptisss commited on 17 days ago

Add Best-of-N rejection sampling + RFT pipeline for value fidelity improvement

f82a9bd
verified

nraptisss commited on 17 days ago